Although behavioral interventions for dementia caregivers have been shown to decrease caregiver stress and burden, there has been little examination of their effects on healthcare costs for patients or caregivers. This lack of information on interventions' impact on healthcare costs, particularly for dementia patients, has likely impeded widespread adoption of successful interventions. The proposed study addresses important knowledge gaps in our understanding of whether a successful behavioral intervention for caregivers has an impact on actual healthcare costs for patients and caregivers and whether those impacts vary based on caregiver and patient characteristics. Our main study goal is to conduct a cost identification analysis to determine if there are healthcare cost savings for dementia patients associated with their caregivers' participation in a successful behavioral intervention to improve their own coping and patient management. We will also examine whether participation in the intervention results in healthcare use and cost savings for caregivers. We propose to integrate patient and caregiver data from two previous studies of successful behavioral interventions for dementia caregivers with Medicare and VA healthcare use and cost data. The first study was the national randomized controlled trial Resources for Enhancing Alzheimer's Caregiver Health (REACH II), conducted from 2000 to 2003, with 642 caregiving dyads from 5 sites. The second study was the national clinical translation of REACH II into the Department of Veterans Affairs healthcare system, REACH VA, with 127 dyads from 29 sites, conducted from 2007-2009. For the REACH II data, half the caregivers received the behavioral intervention designed to assist with their own coping and the management of the patient's dementia related behaviors; the other half received usual care. For each patient, we will examine Medicare healthcare use (e.g., physician visits) and their associated costs for one year before the intervention, for the time during the intervention, and for one year after the intervention. Use and costs will be compared for patients whose caregivers received the intervention vs. those who did not and for caregivers who received the intervention vs. those who did not. For the REACH VA study, all caregivers received the intervention. Both VA and Medicare use and costs will be compared for the patients whose caregivers received the intervention vs. a matched comparison group of other VA patients with dementia. For all groups, we will determine if use and costs are different for different types of patients and caregivers (e.g., those of different races or ethnicities).